Talk:Incidence matrix

what will be the incidence matrix of a null graph.

Incidence Matrix Construction Missing[edit]

There is no concrete example of how to construct an Incidence Matrix. The following link provides an excellent introduction. http://mathworld.wolfram.com/IncidenceMatrix.html The incidence matrix should link back to the Fano plane.

http://math.ucr.edu/home/baez/octonions/node4.html is an excellent introduction to the Fano plane. GeMiJa 16:23, 12 June 2007 (UTC)[reply]

Isn't it a miracle there are no concrete matrices in this article? It is quite queer for me. Gubbubu (talk) 13:24, 12 July 2008 (UTC)[reply]

There are examples here but they don't go into details. See French article for more detailed concrete examples. I don't want to mess up with the English article, I am no native English speaker, so I cowardly leave the task of adding more detailed examples to someone else. --MathsPoetry (talk) 15:35, 18 March 2012 (UTC)[reply]

Same as Seidel Adjacency Matrix?[edit]

Is the incidence matrix for a directed graph the same as the seidel adjacency matrix? 115.129.11.42 (talk) 01:52, 30 July 2009 (UTC)[reply]

No. That article says it has "row and column for each vertex". Columns of incidence matrix correspond to edges, not vertices. X7q (talk) 04:32, 30 July 2009 (UTC)[reply]

Square example[edit]

It bothers me a bit that the example in Incidence matrix#Undirected and directed graphs happens to be a square matrix. Isn't this doing a disservice? If someone is trying to get the basic idea of adjacency and incidence matrices, showing only square matrices as examples of both sounds like a bad idea to me. I would much rather show a more "typical" example of an incidence matrix, with more columns than rows. (With more than 4 columns it would be also easier to see that each column has exactly two ones.) Should we change the graph, or is there a good reason for using this particular example? — Miym (talk) 20:48, 11 September 2009 (UTC)[reply]

True. See the example graphs in the French article, I avoided using the same number of vertices and edges and took the "house-like" graph from the German article. --MathsPoetry (talk) 15:32, 18 March 2012 (UTC)[reply]

Relationship between adjacency and incidence matrices[edit]

The article only mentions the relationship with the adjacency matrix of a line graph; it neglects the (simpler) relationship with the adjacency matrix of the original graph. — Preceding unsigned comment added by 78.45.97.159 (talk) 09:21, 21 February 2012 (UTC)[reply]

Very very true, thanks. I added this relation to the French article about Incidence matrices. Hopefully seomeone does the same on the English Wikipedia. --MathsPoetry (talk) 15:30, 18 March 2012 (UTC)[reply]

Unaccuracy[edit]

The formula A(L(G)) = B(G)^T B(G) - 2 I_q is said in this article to work both with oriented and unoriented graphs.

That is not true, with an oriented graph you end up with coeficients equal to -1 in the matrix, which are not legitimate in an adjacency matrix.

Also, I don't really know whether line graphs of oriented graphs do exist at all. the wikipedia article about line graphs define them only in the context of unoriented graphs.

In the hope that helps, --MathsPoetry (talk) 13:36, 18 March 2012 (UTC)[reply]

fix the bad edit[edit]

some one made a bad edit and messed up the formatting. Please fix this, thanks.